In the above-mentioned swash-plate type compressors a kind wherein a swash-plate chamber itself concurrently functions as a passage for the refrigerant gas is known. This kind of compressors have been, however, problematic in two respects, i.e., (1) unsmooth flowing of refrigerant gas due to a large resistance to the gas flow in the swash-plate chamber, and (2) frequent occurrence of noise in the operation time from engagement between the swash-plate and the piston.
The large resistance in the swash-plate chamber to the gas flow seems to be against normal intuitive thinking, but the refrigerant gas does not flow smoothly, in fact unexpectedly, because of a possible rotational movement of the gas around the swash-plate induced by the high speed rotation thereof.
The operation noise, whose cause was long unknown, has been made clear by a long study in the laboratory by the applicants, who found that it is attributable to a gradual increase of clearance between the engaging members on both sides, that is, the piston on one side and the swash-plate, shoe, etc. on the other side. Although the rotating members such as the swash-plate, shoe, ball, etc. are all cooled sufficiently and universally by the refrigerant gas, the piston, particularly the connecting portion thereof, which is reciprocated in one limited place and cooled only from its inside surface facing the swash-plate chamber by the refrigerant gas can not be fully cooled. Specifically, the pistons distantly located from the refrigerant gas inlet are apt not to be cooled enough.
It has been made clear, as a result of the study, that a certain thermal-expansion difference which takes place between the well cooled portion including the swash-plate, shoe, ball, etc., and the unsatisfactorily cooled portion of the piston causes an increase of the shoe clearance, leading to the noise production.